Detection and partial characterization of proenkephalin mRNA

2 downloads 0 Views 951KB Size Report
ABSTRACT We have used an oligodeoxynucleotide of defined sequence to detect and quantitate proenkephalin mRNA in the poly(A)-containing fraction of RNA ...
Proc. Natd Acad. Sci. USA Vol. 78, No. 9, pp. 5484-5487, September 1981 Biochemistry

Detection and partial characterization of proenkephalin mRNA (onigodeoxynucleotuie/hnyDbrizaion/ci.NAi/sequence determination/specific priming) UELI GUBLER*, DANIEL L. KILPATRICK*, PETER H. SEEBURGt, L. PATRICK GAGE4, AND SIDNEY UDENFRIEND* *Roche Institute of Molecular Biology, Nutley, New Jersey 07110; tDivision of Molecular Biology, Genentech, Inc., South San Francisco, Calibrnia 94080; and

tDepartment of Molecular Genetics, Hoffinann-La Roche, Inc., Nutley, New Jersey 07110 Contributed by Sidney Udenfriend, June 5, 1981

ABSTRACT We have used an oligodeoxynucleotide ofdefined sequence to detect and quantitate proenkephalin mRNA in the poly(A)-containing fraction ofRNA from bovine adrenal medullas. The decahexamer 5'-d(G-G-T-A-G-T-C-C-A-T-C-C-A-C-C-A)-3' was synthesized to be complementary to the codons specifying the amino acid sequence NH2-Trp-Trp-Met-Asp-Tyr-Gln-COOH. This stretch of amino acids occurs in peptide I, one of the intermediates in the biosynthetic pathway of the enkephalins in bovine adrenal medulla. This pathway starts with a precursor (proenkephalin) of about 45 Idlodaltons [Stern, A. S., Jones, B. N., Shively, J. E., Stein, S. & Udenfriend, S. (1981) Proc. NatL AcaeS Sci USA 78, 1962-1966]. The decahexamer hybridized to adrenal poly(A)+RNA and was extended into cDNA with reverse transcriptase (RNA-dependent DNA nucleotidyltransferase). Five main discrete products ranging in size from 115 to 168 nucleotides were observed. The sequences of these extensions were found to be identical over the approximately 70 nucleotides sequenced from their 5' termini and corresponded exactly to the sequence expected from the amino acid sequence of peptide I. These cDNAs and the decahexamer itself hybridized to an adrenal medullary poly(A)+RNA species of about 1500 nucleotides, sufficient in size to code for the proposed proenkephalin. At saturation, approximately 2 fmol of the decahexamer were bound per jzg of mRNA; thus, the proenkephalin mRNA represents about 0.1% of the total poly(A)+RNA population in the tissue.

indicating that the proenkephalin mRNA was present in low abundance. The availability of amino acid-sequence information from the apparent intermediates in the enkephalin pathway made it possible to devise an oligodeoxynucleotide probe of defined sequence. The use of such probes for the detection of low-abundance mRNAs has been documented (3-6). Our probe is complementary to the codons- specifying the amino acid sequence NH2-Trp-Trp-Met-Asp-Tyr-Gln-COOH, which occurs in one of the enkephalin-containing peptides isolated from bovine adrenal medulla, peptide I (Fig. 1). Because of the low degeneracy of the code for five of the amino acids in this sequence, synthesis of only a. single complementary oligodeoxynucleotide was necessary. By including the first nucleotide ofthe Gln-codon (see Fig. 1), a fully defined decahexamer could be devised by the rules outlined in ref. 7. Besides its lack of ambiguity, the length of this probe made it extremely useful in the detection, partial characterization, and quantitation of proenkephalin mRNA.

Studies on the biosynthetic pathway of the opioid pentapeptides [Met]enkephalin and [Leu]enkephalin in bovine adrenal medulla have led to the isolation and partial sequencing of several enkephalin-containing polypeptides (1). These polypeptides are most likely the intermediates in a pathway that starts with a precursor protein called '"proenkephalin," which is possibly as large as 45 kilodaltons (kDal) (1, 2). Proenkephalin is a "multivalent"' protein that contains one [Leu]enkephalin and seven [Metlenkephalin sequences and is a candidate for the primary translation product of the enkephalin pathway. To learn more about this protein, we have begun to characterize its mRNA. The cloning and sequencing of this mRNA should facilitate the elucidation of the protein's primary structure, its processing to intermediates and end products in the biosynthetic pathway, and the regulation of its expression. Initially, we set out to see if we could detect proenkephalin synthesis from total adrenal poly(A)+RNA in an in vitro system. Because the purification of. proenkephalin has not been achieved, an antibody directed against the whole protein is not yet available. Therefore, the in vitro translation assays were monitored with an antibody directed against [Metlenkephalin after treatment of the translational products with proteases as described (2). These assays proved not to be sensitive enough,

MATERIALS AND METHODS Reverse transcriptase (RNA-dependent DNA nucleotidyltransferase) from avian myeloblastosis virus was provided by J. W. Beard (Life Sciences, St. Petersburg, FL). T4 polynucleotide kinase and oligo(dT)-cellulose (type 7) were from PL-Biochemicals. [y-32P]ATP (specific activity, -5000 Ci/mmol; 1 Ci = 3.7 X 1010 becquerels) was purchased from Amersham or ICN. Bovine adrenal glands were obtained within 30 min of slaughter. The cortices were removed, and the medullas immediately were placed in liquid nitrogen for storage. Total RNA was extracted by the method of Hall et ad (8). Poly(A)+RNA was obtained by three passages over oligo(dT)-cellulose (9). The decahexamer 5'-d(G-G-T-A-G-T-C-C-A-T-C-C-A-C-C-A)-3' was synthesized by using the modified triester approach (10). T4 polynucleotide kinase and [y-32P]ATP were used to label its 5' ends (11). This decahexamer was used as a primer for the synthesis of cDNA from adrenal medullary poly(A)+RNA as described (7). The cDNA products were fractionated on 6% (wt/ vol) polyacrylamide/7 M urea slab gels, visualized by autoradiography, eluted, and the sequence was determined by the method of Maxam and Gilbert (11). RNA was sized on agarose gels in the presence of 10 mM CH3HgOH and transferred-onto sheets of nitrocellulose (12, 13). Hybridizations ofthe decahexamer to RNA fixed on nitrocellulose (see Fig. 5) were carried out in solution A [0.6 M sodium chloride/0.06 M sodium citrate/ 0.2% bovine serum albumin/0.2% Ficoll/0.2% polyvinylpyrrolidone/0. 1% NaDodSOjpartially hydrolyzed yeast RNA (2 mg/ml)/sonicated and denatured salmon sperm DNA (150 ,ug/

The publication costs ofthis article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. §1734 solely to indicate this fact.

Abbreviation: kDal, kilodalton(s). § The term multivalent is used to indicate a polypeptide wih many identical functional sequences.

5484

Biochemistry: Gubler et d

Proc. NatL Acad. Sci. USA 78 (1981)

5485

1 5 10 15 20 Ser - Pro-Thr- Leu- Glu-Asp-Glu-His-Lys-GIu-Leu-Gin-Lys-Arg-Tyr-Gly-Gly-Phe- Met- Arg-

25 30 35 39 Arg -Val-Gly - Arg -Pro-Glu-Trp-Trp-Met-Asp-Tyr-GIn-Lys-Arg-Tyr-Gly-Gly-Phe-Leu

FIG. 1. Amino acid sequence of peptide I from ref. 1. The internal amino acid sequence used for devising the probe is underlined.

ml) (14)]. The nitrocellulose filters were then washed with seven changes of solution B for 30 min each. The first five washes were at 300C and the last two were at 35QC. Solution B is 0.6 M sodium chloride/0.06 M sodium citrate/0.1% NaDodSO4partially hydrolyzed yeast RNA (20 jug/ml). Hybridization of cDNA to nitrocellulose-fixed RNA was carried out as described (15), with sodium phosphate and glycine omitted from the buffer. RESULTS Qualitative Detection of Proenkephalin mRNA. Increasing amounts of poly(A)+RNA from bovine adrenal medulla or rat liver were bound to nitrocellulose filters, hybridized with labeled decahexamer and washed extensively. Subsequently, the radioactivity in the hybrids was measured. The labeled decahexamer hybridized in proportion to the filter-bound adrenal poly(A)+RNA (Fig. 2). The small amount of hybridization of the probe to rat liver RNA in these experiments represents background. When rat liver poly(A)+RNA was fractionated on gels and transferred to nitrocellulose, no hybridization of the probe was detected (data not shown). Use of the Decahexamer to Prime cDNA Synthesis from Adrenal Poly(A)+RNA. Under appropriate conditions, the decahexamer should prime the synthesis of proenkephalin-specific cDNA from adrenal poly(A)+RNA. Primer labeled at the 5'-end and RNA were hybridized in 0.1 M NaCl, with subsequent assembly of the reverse transcription reaction mixture (7). cDNA synthesis was then allowed to proceed for 90 min at 380C. The products were sized on 6% polyacrylamide/7 M urea

gels (Fig. 3). In five different experiments with two batches of independently isolated mRNA, only the five bands indicated in Fig. 3 were reproducibly obtained over a less distinct background. Each ofthefive bands represents the extension of about 0.03% of the total amount of decahexamer used. Sequencing the Specific cDNA Extensions. If the priming events involved in the cDNA synthesis had occurred at the expected position on the proenkephalin mRNA, the DNA sequence should be the one that codes for the known amino acid sequence ofpeptide I. Several large-scale cDNA syntheses (250 1.d each) were carried out, and each ofthe five cDNA bands was eluted from the gels and partially sequenced. Over the 70 nucleotides sequenced thus far from their 5'-termini, the extensions were found to be identical for all five bands. Fig. 4 comcDNA

603

310 281-i. 1 271 234 194

*::.0 -1s 168

"~6

-151 143

o0 X

118

4-

CU

2

__

-

0.2

0.4 0.6 0.8 mRNA per filter, Axg

72

FIG. 2. Hybridization of labeled decahexamer to nitrocellulose-

bound poly(A)+RNA from bovine adrenal medulla and rat liver. A total of 26 filters (duplicate filters carrying increasing amounts of RNA)

were prewashed in 13 ml of solution A for 4 hr at 30TC. Nine ml of solution A was then removed, 5 pmol of decahexamer was added (specific activity, 3.7 x 106 cpm/pmol), and the mixture was heated to 650C for 10 min and then transferred to hybridization temperature (300C) for 14 hr. The filters were washed in 26 ml of solution B, dried, and counted. Filter retention of the RNA was monitored by incubating labeled adrenal poly(A)+RNA on parallel filters under the same conditions. Experimental points have been corrected for blank values and the RNA retention efficiency (20% in this particular experiment). *, adrenal poly(A)+RNA; o, rat liver poly(A)+RNA.

FIG. 3. Size analysis of cDNA primed with decahexamer. Prehybridization of primer and mRNA (0.5 pmolflg) was as described (7). cDNA was synthesized in 50 mM Tris HCl, pH 8.3/15 mM MgCl258 mM NaCl/5 mM DDT/0.5 mM deoxynucleoside triphosphates/ poly(A)+RNA (170 gg/ml)/reverse transcriptase (0.75 units/pg of RNA) for 90 min at 38TC. The products of the reaction were phenol extracted, the template RNA was hydrolyzed in 0.3 M NaOH (9), and the cDNA was analyzed on a 6% polyacrylamide/7 M urea gel (40 V/ cm for 60 min). Bands were visualized by short-time autoradiography (right-side arrows). Kinase-labeled HaeIH-digested 4X174 fragments were included as size standards (left-side arrows) expressed in no. of nucleotides.

5486

Biochemistry: Gubler et al.

Proc. Nad Acad. Sci. USA 78 (1981) .-[Met]

Peptide I

NH2 -

Enkephalin -

-Argi yr-Gly-Gly-Phe-Met T

Expected cDNA sequence

3' GCN

Observed cDNA sequence

Arg-Arg-Val-Gly-Arg-Pro-Glu Trp-Trp-Met-Asp-TyrGCNGCN

T -CAN-CCN- cC-GGN-CT C-TCGCN

T-CCN-CCN G

n --- COOH

Primer. ACC-ACC-TAC-CTG-ATG-

5

3'---GCT-ATG-CCC-CCA-AAA-TAC-TCT-TCT-CAC-CCA-GCA-GGC-CTC-ACC-ACC- - 5'

FIG. 4. Partial sequence of the cDNA extensions described in Fig. 3. The codons for the partial amino acid sequence from peptide I (top line) and their complementary sequence (middle line) were derived. About 70 nucleotides of the cDNAs were sequenced and found to correspond to the sequence expected on the basis of peptide I. Of that sequence, 39 nucleotides are shown on the bottom line.

pares parts of the experimentally determined sequence of the extensions with the sequence expected on the basis of peptide I. Clearly, the two sequences match up exactly, indicating that the cDNA products obtained were primed from the expected region in the proenkephalin mRNA. These different cDNA molecules most probably reflect initiations at the same site on the proenkephalin mRNA. They differ in length because of specific transcription termination events that might be due to secondary structure of the mRNA. The Size of Proenkephalin mRNA. The specificity of the reverse transcription having been established, both decahexamer and cDNA proved useful as probes in the sizing of proenA

B

Origin

28 S

-

23 S

M.

xX -

I

I

2

4

I

I

I

6

18 S

16 S

kephalin mRNA. Total poly(A)+RNA from adrenal medullas was fractionated on denaturing agarose gels and blotted onto nitrocellulose. Parallel lanes were hybridized with either the decahexamer or a mixture ofthe five cDNAs prepared as described. Both the decahexamer and the cDNAs hybridized to a discrete poly(A)+RNA ofthe same size (Fig. 5). Standards run in parallel lanes indicated that this RNA is about 1500 nucleotides in length. The Abundance of Proenkephalin mRNA in Bovine Adrenal Medulla. Because it is likely that the sequence used to devise the decahexamer occurs only once per proenkephalin mRNA molecule, a saturation hybridization study should yield useful information about the abundance of the specific mRNA in the total population of adrenal medullary poly(A)+RNA. Constant amounts of RNA were bound to filters and hybridized with increasing amounts oflabeled decahexamer. The filters were then processed. Fig. 6 shows a plot of R (input decahexamer) versus R/DR (input decahexamer/amount of decahexamer hybridized). The reciprocal of the slope of this linear plot yields the saturation value, Hsat (16). This value was estimated to be 2 fmol of decahexamer hybridized at saturation per ,ug of mRNA. By using the estimated size of 1500 nucleotides for the proenkephalin mRNA (Fig. 5), it was calculated that the mRNA in question represents about 0.1% of the poly(A)+RNA, or about 0.003% of the total RNA of the tissue.

_

FIG. 5. Hybridization of decahexamer or cDNA to adrenal medullary poly(A)+RNA. Ten tg of poly(A)+RNA was fractionated on parallel lanes on a 1.4% (wt/vol) agarose gel in the presence of 10mM CH3HgOH for 16 hr at 3 V/cm. The RNA was transferred onto nitrocellulose sheets. Adjacent lanes were hybridized with decahexamer [2 pmol (specific activity, 5.7 x 106 cpm/pmol) in 3 ml of solution A for 17 hr at 30°C] or with cDNA [12,000 cpm for 12 hr at 42°C, as described (15)]. Filters were washed, dried, and autoradiographed at -70°C. Size markers were Escherichia coli rRNA and bovine rRNA, run in parallel lanes and visualized by staining with ethidium bromide prior to transfer. Lanes: A, decahexamer; B, cDNA.

4

1: 2

8 6 x fmol 10-3 R,

10

FIG. 6. Saturation hybridization of labeled decahexamer to adrenal medullary poly(A)+RNA. R, input decahexamer; R/DR input decahexamer/amount of decahexamer hybridized. Nitrocellulose filters were loaded with4 ,ug of RNA. Four filters for each point (3 RNA filters plus one blank) were prewashed in 1.3 ml of solution A for 4 hr at 30°C; 0.8 ml of solution A was then removed, the indicated amounts of decahexamer were added (specific activity, 9.5 x 106 cpm/pmol) and the mixes were heated to 60°C for 10 min. Hybridization was carried out for 18 hr at 30°C. Washes were done in 5 ml of solution B. Filter retention was quantitated by monitoring loss of labeled reoviral mRNA from parallel filters. Experimental points have been corrected for blank values (usually less than 10% of the experimental values) and the RNA retention efficiency (30% in this particular experiment).

Biochemistry: Gubler et d DISCUSSION The experiments illustrate the usefulness of short DNA primers of defined sequence for detecting a specific mRNA present in low abundance. This approach is rapid, specific, and highly sensitive. In our instance, a specific mRNA that represents 0.003% ofthe total RNA from a tissue was unequivocally detected. This species ofbovine adrenal medullary poly(A)+RNA fulfills stringent criteria of identification as the proenkephalin mRNA. In particular, it contains sequences that were predicted on the basis of the amino acid sequence of one of the intermediate enkephalin-containing polypeptides. Moreover, its size (1500 nucleotides) is sufficient to code for the proposed multivalent proenkephalin (40-50 kDal). A report has appeared concerning the use of in vitro translation to characterize bovine adrenal proenkephalin mRNA (17). There are several differences between our data and this study. Two forms of proenkephalin were reported, one as large as 110 kDal; the mRNA was assigned a size of about 21 S upon fractionation under nondenaturing conditions. The appearance of several cDNA products of different lengths but with identical 5'terminal sequences deserves further comment. The fact that they were of different lengths most likely reflects specific transcription termination events due to secondary structure in the mRNA. As explained earlier, the five cDNAs were most probably primed at the same site on the mRNA. Data from the initial characterization of proenkephalin (2) make it unlikely that the primer sequence and the 70 bases of sequence determined for each cDNA occur five times per mRNA molecule. To our knowledge, multiple identical sequences ofthis length appearing in a single gene have not been reported so far. The presence of a series of enkephalin-containing polypeptides in adrenal medulla, ranging in size from proenkephalin (40-50 kDal) to free enkephalins (0.5 kDal), suggests that they are related through a biosynthetic pathway (1). Appreciable evidence of a circumstantial or preliminary nature (18, 19) supports such a pathway. The present studies are quite conclusive: a probe based on a sequence in peptide I, one of the intermediate-sized enkephalin-containingpolypeptides, primes cDNA that contains nucleotide sequences deduced from peptide I. Furthermore, this cDNA hybridizes to a single mRNA that is sufficient in length to code for proenkephalin. These findings represent the most direct evidence for the fact that proenke-

Proc. NatL Acad. Sci. USA 78 (1981)

5487

phalin is the precursor of at least one of the adrenal enkephalincontaining polypeptides, peptide I. Note Added in Proof. Since this manuscript was submitted for publication, we have cloned and sequenced most of the proenkephalin mRNA.

We are grateful to Roberto Crea from Genentech Inc., San Francisco, for making the primers available to us. Special thanks go to Mike Tocci, Peter Lomedico, and John Monahan for their help and advice and to Diane Andriola for preparing the manuscript. 1. Stem, A. S., Jones, B. N., Shively, J. E., Stein, S. & Udenfriend, S. (1981) Proc. Natl Acad. Sci. USA 78, 1962-1966. 2. Lewis, R. V., Stem, A. S., Kimura, S., Rossier, J., Stein, S. & Udenfriend, S. (1980) Science 208, 1459-1461. 3. Noyes, B. E., Mevarech, M., Stein, R. & Agarwal, K. L. (1979) Proc. Natl Acad. Sci. USA 76, 1770-1774. 4. Houghton, M., Stewart, A. G., Doel, S. M., Emtage, J. S., Eaton, M. A. W., Smith, J. C., Patel, T. P., Lewis, H. M., Porter, A. G., Birch, J. R., Cartwright, T. & Carey, N. H. (1980) Nucleic Acids Res. 8, 1913-1931. 5. Hudson, P., Haley, J., Cronk, M., Shine, J. & Niall, H. (1981) Nature (London) 291, 127-131. 6. Sood, A. K., Pereira, D. & Weissmann, S. M. (1981) Proc. NatL Acad. Sci. USA 78, 616-620. 7. Agarwal, K. L., Brunstedt, J. & Noyes, B. E. (1981) J. Biol Chem. 256, 1023-1028. 8. Hall, L., Craig, R. K. & Campbell, P. N. (1979) Nature (London) 277, 54-56. 9. Norgard, M. V., Tocci, M. J. & Monahan, J. J. (1980) J. Biol Chem. 255, 7665-7672. 10. Crea, R., Kraszewski, A., Hirose, T. & Itakura, K. (1978) Proc. Natl Acad. Sci. USA 75, 5765-5769. 11. Maxam, A. M. & Gilbert, W. (1980) Methods EnzymoL 65, 499-559. 12. Bailey, J. M. & Davidson, N. (1976) AnaL Biochem. 70, 75-85. 13. Thomas, P. S. (1980) Proc. Nati Acad. Sci. USA 77, 5201-5205. 14. Mevarech, M., Noyes, B. E. & Agarwal, K. L. (1979) J. BioL Chem. 254, 7472-7475. 15. Alwine, J. C., Kemp, D. J. & Stark, G. R. (1977) Proc. Natl Acad. Sci. USA 74, 5350-5354. 16. Bishop, J. 0. (1972) Acta Endocrinol Suppl 168, 247-276. 17. Dandekar, S. & Sabol, S. L. (1981) Fed. Proc. Fed. Am. Soc. Exp. Biol 40, 1642 (abstr.). 18. Lewis, R. V., Stern, A. S., Kilpatrick, D. L., Gerber, L. D., Rossier, J., Stein, S. & Udenfriend, S. (1981) J. Neurosci. 1, 80-82. 19. Rossier, J., Trifaro, J. M., Lewis, R. V., Lee, R. W. H., Stern, A., Kimura, S., Stein, S. & Udenfriend, S. (1980) Proc. Natl Acad. Sci. USA 77, 6889-6891.